Download pdf - DOPAS WP3 Deliverable D3 - IGD-TP · This report DOPAS WP3 Deliverable D3.7 first gives an overview of the laboratory work, carried out in order to define the most suited clayish

Dopas Deliverable D3.7 Dissemination RE 1

DOPAS WP3 Deliverable D3.7“Test report on FSS metric clayish material emplacement tests with

clayish material definition and laboratory work on its performance”

Start date of the project: September 2012 Duration: 48 months

Project co-funded by the European Commission under the Euratom Research and TrainingProgramme on Nuclear Energy within the 7th Framework Programme (2007-2013)

Dissemination level

PU Public X

RE Restricted to a group specified by the partners of theProject, including EC

CO Confidential, only for DOPAS partners

Grant Agreement number: 323273Authors: Jean-Michel Bosgiraud

Régis Foin(Andra)

Date of preparation: 08 December 2016Version status: A

Dopas Deliverable D3.7 Dissemination PU 2

Important Note:The three following DOPAS WP3 Deliverables (as identified below according to the initialDOPAS Description of Work), all related to the FSS experiment, have been merged in onesingle document:

· DOPAS Deliverable D3.3: “Report on clayish material definition for FSS”,

· DOPAS Deliverable D3.5: “Lab test report on the performance of the FSS clayishmaterial”,

· DOPAS Deliverable D3.7: “Test report on FSS clayish material metric coreemplacement”.

The new global document is titled:

· DOPAS WP3 Deliverable D3.7: “Test report on FSS metric clayish materialemplacement tests with clayish material definition and laboratory work on its performance”.

History Chart

Revision Document name Partner Date

Draft 1 DOPAS WP3 Deliverable D3.7: “Test report on FSS metricclayish material emplacement tests with clayish materialdefinition and laboratory work on its performance”

Andra 18.12.2013

Draft 2 DOPAS WP3 Deliverable D3.7: “Test report on FSS metricclayish material emplacement tests with clayish materialdefinition and laboratory work on its performance”

Andra 14.03.2014

Version A DOPAS WP3 Deliverable D3.7: “Test report on FSS metricclayish material emplacement tests with clayish materialdefinition and laboratory work on its performance”

Andra 08.12.2016

REVIEW/OTHER COMMENTS:The report was internally reviewed by Andra and later submitted in version A for approvalby the DOPAS Coordinator (Johanna Hansen for POSIVA).

APPROVED FOR SUBMISSION:Johanna Hansen (POSIVA) 9.12.2016

Scope Deliverable n°D3.7 (WP3) Version: A

Type/No. Report D3.7 Total pages 34

Title DOPAS WP3 Deliverable D3.7: “Test report onFSS metric clayish material emplacement testswith clayish material definition and laboratorywork on its performance”

Articles: 1 + 8


Executive Summary

In DOPAS, Work Package 3 (WP3) is related to the construction of large scale demonstratorsof seals and plugs. FSS (Full-scale Seal) is the “seal industrial demonstrator” (prototype)designed and built by Andra with the scientific help of NAGRA, while other WMO’s (i.e. theDOPAS partners) are building or have built or will build their own prototypes (namelyDOMPLU, POPLU, EPSP, ELSA).This report DOPAS WP3 Deliverable D3.7 first gives an overview of the laboratory work,carried out in order to define the most suited clayish (bentonitic) material for the needs ofAndra’s FSS experiment, i.e. building a swelling clay core inside a test box (drift model),with the relevant “emplaced dry density” value and the adequate permeability performanceand the most suited swelling pressure. The material used is a mix of WH2 (a bentonite brandequivalent to MX80) based pellets and powder (made of crushed pellets). It is also tested inthe REM Experiment (a 10-30 year long hydration test in a metric test cell), coming as acomplement to FSS, in order to assess the material performance at a scale which is ten timesthat of the tests carried out in lab (at a decimetric scale, over a few months) and provide datato further modelling.The document includes an overview of the main experimental phases (illustrated by photos,graphs and tables) implemented to determine the right formulation of WH2 bentonite pelletsand powder admixture, first used for the bentonite material emplacement tests at a metricscale, preceding the full scale swelling clay core construction activities.

The formulation selected is described and its characterization provided. The metricemplacement tests are also presented with their outcomes. They have an impact on the finalbentonitic material specification effectively considered for the FSS swelling clay coreconstruction, as well as on the design of the full-scale backfilling machine to be used in FSS.

Links to previous and future Andra’s FSS specific (or more general DOPAS) deliverables aregiven. The global FSS construction and investigation timeline is summarized in the tablebelow:

Period Activity

August 2012 Beginning of studies

November 2012 – June 2013 Drift model (aka “Test Box”) construction

August 2012 – July 2013 Low-pH SCC concrete and shotcrete mix development

August 2012 – April 2014 Development of bentonitic materials and methodology to emplacethe swelling clay core mix

July 2013 Low-pH SCC “upstream” containment wall construction

August 2014 Swelling clay core construction

September 2014 Low-pH shotcrete “ downstream” containment wall construction

October 2014 – July 2015 Scientific investigations

August 2015 – December 2015 “Investigation Dismantling” followed by a complete deconstructionof FSS and release of test site to landlord


List of Acronyms – AbbreviationsThis list of acronyms is generic. It concerns entities, activities, concepts, equipment andmaterials; some of which are Andra specific in the context of the FSS experiment.

ASN: Autorité de Sûreté Nucléaire (French Nuclear Authority).CIGEO: Centre Industriel de Stockage Géologique (Industrial Geological Repository,

AKA Cigéo).CNE: Commission nationale d’évaluation (National Assessment Board).

CP: Crushed Pellets (bentonite powder made of)DGR: Deep Geological Repository (see also GDF)

DOPAS: Full-scale Demonstration of Plugs and Seals (Name of EC Project on Seals).EBS: Engineered Barrier System.

EC: European Commission.EDZ: Excavation damaged zone.

ESDRED: Engineering Studies and Demonstration of Repository Designs (name of aprevious EC supported Project).

FSS: Full-scale Seal.GDF: Geological Disposal Facility (see also DGR).

GME: Groupement momentané d’entreprises (FSS General Contractor, Consortiumof companies).

HLW: High-level Waste.IAEA: International Atomic Energy Agency.

ID: Internal diameter.ILW: (Long Live) Intermediate-level Waste (also LL-ILW).

IRSN: Institut de Recherche sur la Sûreté Nucléaire (Expert Organization andtechnical support to ASN).

LECBA: Laboratoire d’études et caractérisation des bétons et argiles (Laboratory forstudy and characterization of concrete and clay materials) – member of GME.

LLW: Low-level Waste.MPC: Member of GME – vendor of bentonite based products.

OD: Outside diameter.R&D: Research and Development.

SCC: Self-compacting concrete.URL: Underground research laboratory (Bure is the French URL).

WMO: Waste Management Organization.WP: Work Package.


List of DOPAS Project Partners

The 14 partners from 8 different countries in the DOPAS Project are listed below. In theremainder of this report each partner is referred to as indicated:

Posiva Posiva Oy Finland

Andra Agence nationale pour la gestion des déchets radioactifs FranceDBETEC DBE TECHNOLOGY GmbH Germany

GRSGmbH

Gesellschaft für Anlagen- und Reaktorsicherheit Germany

Nagra Die Nationale Genossenschaft für die LagerungRadioaktiver Abfälle

Switzerland

RWM Radioactive Waste Management Limited UKSÚRAO Správa Úložišť Radioaktivních Odpadu (Radioactive

Waste Repository Authority – RAWRA)Czech Republic

SKB Svensk Kärnbränslehantering AB Sweden

CTU Czech Technical University Czech RepublicNRG Nuclear Research and Consultancy Group NetherlandsGSL Galson Sciences Limited UK

BTECH B+ Tech Oy FinlandVTT Teknologian Tutkimuskeskus VTT Oy (VTT Technical

Research Centre of Finland Ltd)Finland

UJV Ustav Jaderneho Vyzkumu (Nuclear Research Institute) Czech Republic


Glossary of scientific formulations/equations

Term Definition Formula in French

M (g) Mass of mixture

Md (g) Mass of dry mixture

MW (g) Mass of water in mixture

PC (MPa) Compacting pressure

PS (MPa) Swelling pressure

RH (%) Relative Humidity (measured) Humidité relative (HR)

k (m/s)Permeability of saturatedmaterial or hydraulicconductivity

Taux devide

Percentage of voids betweengrains, pellets or powder totalVolume

grainsintervidesvolume=

W (%) Water content (measured at105°C) secmatériaumasse

eaumasseW =

rW (g/cm3) Water density (specificgravity) 1.0 g/cm3

r (g/cm3) Bulk density (specific gravity) totalvolumetotalemasseρ =

rd (g/cm3) Dry density (dry specificgravity) W1

1ρmatériautotalvolumematériausèchemasseρ d +

´==

rred (g/cm3) Emplaced dry density (withvolume increase)

VVΔ1

1ρfinalvolumematériauxsèchemasseρ dred

+´==

rS (g/cm3) Dry density of solids in clay 2,78 g/cm3 for WH2

h (%) Porositys

d

ρρ1

totalvolumeporesvolumeη -==

Sr (%) Saturation rate ( )[ ]÷øö

çèæ -

»´-+´

´´=

sd

ws

sr

ρ1

ρ1

Wρρ1Wρ

ρρWS


Table of ContentsExecutive Summary .................................................................................................................... 3List of DOPAS Project Partners .................................................................................................. 5Glossary of scientific formulations/equations .............................................................................. 61. National Context for the FSS Experiment ............................................................................. 82. FSS Design Basis and Link to the Cigéo Reference Design Basis ......................................... 9

2.1 FSS Design basis ....................................................................................................... 92.2 FSS Test Box Design and construction .................................................................... 10

3. Work methodology used for the FSS clayish material definition......................................... 123.1 Context .................................................................................................................... 123.2 Specifications .......................................................................................................... 123.3 Discussion ............................................................................................................... 133.4 Methodology ........................................................................................................... 14

4. Preparatory work for the FSS clayish material definition .................................................... 174.1 Materials used in the lab tests .................................................................................. 174.2 Initial characterization of materials .......................................................................... 174.3 Description of test equipment .................................................................................. 194.4 Tests with bentonite pellets only .............................................................................. 214.5 Tests with binary mixtures ....................................................................................... 224.6 Tests with the real components ................................................................................ 234.7 Selection of the bentonite formulation ..................................................................... 24

5. Mock-up testing and emplacement machine design ............................................................ 255.1 Introduction ............................................................................................................. 255.2 Outcomes ................................................................................................................ 27

6. Characterization of the bentonite mix ................................................................................. 306.1 Introduction ............................................................................................................. 306.2 Experimental protocol and results ............................................................................ 30

7. Conclusion ......................................................................................................................... 338. References ......................................................................................................................... 34


1. National Context for the FSS Experiment

In France, the repository host rock is the 155-million-year-old Callovo-Oxfordian induratedclayish formation, which lies in the east of the Parisian Basin. The repository project isreferred to as Cigéo. The disposal reference inventory includes Long-Lived Intermediate-level waste (LL-ILW or ILW) from operation, maintenance and decommissioning of nuclearfacilities and HLW from spent fuel reprocessing. The waste will be disposed of in physicallyseparated disposal zones: one for ILW and one for HLW. The repository’s primary functionis to isolate the waste from human activities at the surface and its second function is toconfine radioactive substance and control the transfer pathways which may in the long termbring radionuclides into contact with humans and the environment (ANDRA, 2013). Theprincipal contribution of the seals in Andra’s concept is to provide the second function.

The ILW disposal zone includes several tens of large-diameter disposal vaults, each about500m long. Vault concrete lining and disposal containers provide a cementitious (buffer)environment for the ILW waste. The gaps and voids between waste packages and vault liningcould be left empty or backfilled with cementitious material or neutral filler (e.g. sand).

In the French concept, seals are defined as hydraulic components for closure of largediameter (several meters) underground installations and infrastructure components such asshafts, ramps, drifts1 and ILW disposal vaults. Each seal consists of a swelling clay core(EBS) and concrete containment walls. The conceptual design of drift and ILW disposal vaultseals is the same. The location of seals in the planned Cigéo repository is shown in Figure1.1.

Figure 1.1: Location of the seals (red dots) in the Cigéo concept

1 Drifts are horizontal tunnels, whereas ramps are inclined tunnels, also called declines.


2. FSS Design Basis and Link to the Cigéo Reference Design Basis

The FSS experiment is a full-scale technical demonstration of construction feasibility for adrift and ILW disposal vault seal, being carried out in a warehouse in Saint- Dizier, which isclose to the French URL at Bure.The FSS test calls for a large excavation, with a significant length and a considerable amountof equipment and materials mobilized and emplaced. The Bure URL is essentially aqualification facility, in which the logistical means are somehow limited (transport means,number of people admitted underground, geometry restrictions for large pieces ofequipment,…). Moreover the Bure URL is busy with various other experiments which cannotbe conducted concurrently with large scale experiments such as FSS.For that reason, and for standalone reasons, like global experimental costs, global scheduleconstraints and needs for “investigation” dismantling and observations, it was decided to gofor a surface facility, instead of working underground. The Saint-Dizier site was proposed bythe Contractor (GME) in charge of the FSS test, and accepted by Andra, since the vicinity ofBure (30km), the height of warehouse (more than 10m of free gap under the roof frame), andthe possibility of air parameters control were in line with the experiment technicalspecifications.

2.1 FSS Design basis

The FSS test is part of a wide-ranging programme of R&D and demonstrator experimentsthat was established by Andra in response to the discussions with ASN and the FrenchNational Assessment Board (CNE) in 2009, during which it has been noted that seals, and inparticular drift and ILW disposal vault seals, required some type of industrial and scientificdemonstration in order to achieve licensing authorisation.As a result, R&D studies and demonstration tests have been launched to assess the technicalfeasibility and to develop the post-closure requirements of seals in the repository. Those testscover the performance and constructability issues. FSS belongs to this last category.

The main objective of the FSS test is to develop confidence in, and to demonstrate, thetechnical feasibility of constructing a full-scale drift (or ILW disposal vault) seal. Technicalfeasibility includes demonstrating the ability of the approach used to emplace the clay to besuitable for filling recesses (breakouts) in the clay host rock, and also the capacity to buildlarge low pH concrete containment walls with satisfactory mechanical properties (shotcreteand SCC technologies are tested and compared).

The FSS test is focused on the “construction only” of the seal, and the swelling clay will notbe saturated or otherwise pressurised. The “REM test” which is a saturation test at a metricscale over some 10 to 30 years, will provide valuable information on the way the bentoniticmix developed in FSS will behave with its progressive saturation and enable a bettermodelling of a seal behaviour at large scale (cf. DOPAS Deliverable D4.2 “Report onbentonite saturation test”, Conil et al.2015).

The FSS design basis is described more detailed in different Deliverables of DOPAS WP2.The respective Cigéo seal design basis and that of the FSS experiment are compared andjustified in the DOPAS WP2 Deliverable D2.1 “Design Bases and Criteria” (White etal.2014), and summarized in D2.4 WP2 “Final Report” (DOPAS, 2016).

The conceptual design of the FSS experiment is otherwise illustrated in Figure 2.1 below.


Figure 2.1: Conceptual design of the ANDRA FSS test.

The main difference between the Cigéo reference and FSS design bases for the Andra driftseal is the length of the seal. The real seal underground (in fact, each component: swellingcore or containment walls) will be longer than the seal considered in the FSS experiment. Thediameter(s) should stay close to what is effectively implemented in FSS.The FSS test box (cf. Figure 2.2) is some 7.6m ID and 36m long. The drift concrete liner(70cm thick) and the formation break outs (recesses) likely to be generated by the drift liningdeposition (up to 1m depth at the liner extrados) are simulated in the construction of the driftmodel.Representative underground ambient conditions (temperature around 18-30 °C, hygrometrybetween 50% and 75%), have to be maintained within the drift model.Low pH cast-concrete/shotcrete 5m long containment walls close the volume of the swellingcore, on both sides. The bentonite swelling core length is some 14m.

2.2 FSS Test Box Design and construction

The FSS Test Box design was elaborated between July and October 2012. The Test Boxworkshop drawings were also supplied during that period on the basis of the FSS Test Boxconcept (as specified by Andra), and (following finite element modelling and dimensioning)derived from the schematic presented in Figure 2.2.

The test box design is documented in the DOPAS WP3 Deliverable D3.2 “FSS Tunnel modeldesign report” (Bosgiraud et al.2013).


Figure 2.2: 3D schematic of the FSS test box

The construction of the FSS Test Box followed. It was started in October 2012 and wascompleted in May 2013. The Figure 2.3 shows the Test Box (drift model), commissioned inJune 2013, as “experiment ready”. The FSS Test Box construction story is documented in theDOPAS WP3 Deliverable D3.10 “FSS Drift Model Construction Report” (Bosgiraud etal.2014).

The construction in July 2013 of the upstream low pH SCC containment wall was then madepossible (the wall construction story is detailed in the DOPAS WP3 Deliverable D3.11“Report on FSS cast concrete plug construction”, Bosgiraud et al.2014). The completion ofthe low pH SCC containment wall was an operational prerequisite of the swelling clay coreconstruction.

Additionally outcomes of the metric clay core emplacement tests were needed, while theyhad an impact on the design of the full-scale bentonitic mix emplacement machine.

Figure 2.3: The FSS Test Box ready for seal construction


3. Work methodology used for the FSS clayish material definition

3.1 Context

The swelling clay core is the most critical component of the seal construction works. Theclayish material constitutive of the core must comply with the following criteria (cf.Deliverable D2.1 “Design Bases and Criteria”):

· The swelling pressure must be sufficient to progressively fill the technical voids betweenthe granular material (powder and pellets),

· The swelling pressure must be sufficient to provide an efficient and permanent hydrauliccontact with the drift walls (concrete liner or argillites), to reduce or minimize the EDZ(by facilitating self-sealing), without exceeding the natural geological constraintsprevailing in the host formation. Ideally, the bentonitic material swelling pressure valueshould be as close as possible to (but not exceeding) 7 MPa.

· The saturated material permeability value should not exceed 10-11m/s.The low pH concrete containment walls erected at both ends of the core are necessary tocontain the bentonite mix during its swelling phase and keep the swelling pressure at its peak.

The use of low pH concrete helps to (i) reduce the thermal load generated by the concretehydration, (ii) limit the effect of the alkaline plume generated by the cementitious water(leachate produced at time of the progressive host rock resaturation, following closure of theunderground openings) on the swelling clay and the surrounding argillites, and (iii) increasethe concrete durability in the underground environment.

3.2 Specifications

The bentonite specifications contained in Andra’s FSS “Scope of Work” (i.e. the technicaland contractual document executed by Andra and the GME) are explicit:

· The bentonitic material must be a “Wyoming type sodic Montmorillonite”, under thebrand MX80 or WH2,

· The swelling core must be made of a granular admixture of pulverulent materials (such aspellets and powder),

· The swelling core must be emplaced in the test box with industrial means similar to thoselikely to be employed in a Cigéo drift, at a 500m depth, at time of seal construction (driftor disposal vault closure),

· Similarly, the swelling core commissioning must be implemented with means andmethods transposable to Cigéo at time of seal construction activities. These means andmethods must then be qualified,

· The performances expected from the emplaced material (at saturation state) are to befulfilled everywhere in the core: swelling pressure equal (or as close as possible) to 7MPa, permeability equal to (or less than) 10-11 m/s.


All these specifications were integrated in the bentonitic material definition andcharacterization process described here after.

3.3 Discussion

The return of experience (from previous experiments and laboratories tests) shows that themost challenging parameter to fulfill is the swelling pressure value, set as close as possible to7 MPa. This swelling pressure value is corresponding to a pre-established specific gravity(dry emplaced density of the swelling material) value. This value is empirically determinedand is specific of the clayish (bentonitic) material concerned.

Data available show that, for a “Wyoming type” bentonite, this density is around 1.56 g/cm3

for pre-compacted blocks swelling in a quasi-constant volume. For a granular material, with ahigher porosity, the effective emplaced dry specific gravity is higher, near 1.60 g/cm3. Theinter-pellets voids/spaces are critical to reach such a value; hence various pulverulentadmixtures (formulations) must be studied to minimize those voids (the purpose is to obtainthe best possible compacity at time of admixture emplacement).

The other specifications are easier to reach. Experience shows that a sodic “Montmorillonite”such as the WH2 type has a permeability value (at saturation state) around 10-13 m/s, for aneffective emplaced density of 1.60 g/cm3 coherent with a 7 MPa swelling pressure.

The selection of the bentonite materials for the swelling clay core was undertaken in thefollowing steps:

· First, a decision was made to adopt a pellet-based admixture rather than use ofpre-compacted bentonite blocks. This choice is motivated by Andra’s belief that the useof pre-compacted blocks (bricks) is not a choice technically commensurate with the needfor emplacing very large volumes of swelling clay in Cigéo in an industrial way.

· Second, laboratory testing of bentonite pellet and powder mixtures was undertaken inparallel with manufacturing tests to identify the appropriate pellet and powder mixture,and initial water content.

· Third, mock-up testing (at a metric scale) and desk-based design work was used to testand develop the design of the bentonite emplacement method.

Note about Adoption of a Pellet-based System:

Andra selected a pellet-based system instead of pre-compacted bentonite blocks because thissolution is considered by Andra to be a more efficient industrial method of implementationfor significant quantities of material. The method is similar to that proposed by NAGRA foremplacement of bentonite buffer materials in its FE experiment at Mont-Terri (cf. EC fundedFP7 Project LUCOEX). In the case of FSS (and of the large seals envisaged by Andra for theCigéo repository), the pellets/powder can be emplaced by conveyor systems while the blocksare to be positioned by human action or robots at a much lower emplacement speed.Furthermore, the erection of a wall of blocks also raises the issue of its stability, since theblocks are not assembled with a mortar.


3.4 Methodology

The methodology which was implemented for the definition and characterization of the bestpossible bentonitic mixture is defined in a 10 step process described below:

Step 1: Supply of material studied and characterized

· It was decided that the supply of bentonite (under the brand name “WH2”) would beordered at a single Wyoming mine site and delivered in one single batch (some 1500metric tons), in order to work with the most homogeneous material possible in terms ofsmectite content. The composition of the admixture was also pre-determined byexperience with a decision to dry mix WH2 bentonite powder and pellets.

· The conditioning could be in pellets OD 32 mm, or/and in pellets OD 7 mm, and inpowder (made or not of crushed pellets) with grains sized between 0.1 and 2 mm/5 mm.

· Supply of steel balls (simulating the different pellets sizes and arrangements) was alsoordered.

Step 2: Manufacturing of test equipment

· Some preliminary pellet manufacturing tests (with an existing compacting machine) werecarried out to optimize the compacting pressure and other production parameters neededto design and later manufacture the final pellet production machine,

· A polycarbonate test cell (ID 390mm, H 500 mm) was ordered, to be used later for thevisualizing and weighing of the various pellets and powder assemblies (Figure 3.1),

· A test cell (ID 120 mm) for the measurement of the bentonitic material swelling pressureand permeability was also ordered and delivered (Figure 3.2).

Step 3: Characterization of the admixture components

The following physical measurements were carried out:

· Geometry of pellets,· Bulk density (specific gravity),· Water content,· Emplaced density,· Powder granulometry.Step 4: Sampling of polycarbonate test cell (ID 390mm, H 500 mm)

The following physical measurements were carried out:

· Geometry and weight of test cell (empty),· Weighing of cell filled with steel balls,· Weighing of cell filled with preliminary pellets.


Step 5: Study of various admixture combinations in the polycarbonate cell

The following work plan was carried out (in order to obtain the best possible compacity,hence the best possible emplaced dry density):

· Optimizing of binary mixes,· Optimizing of ternary mixes,· Tests of admixtures with more than 3 components.

Figure 3.1: Polycarbonate test cell (ID 390mm, H 500 mm) used for the visualizing andweighing of the various pellets and powder assemblies

Figure 3.2: Test Cell (ID 120 mm) for measurement of swelling pressure and permeability


Step 6: Study of various storing, handling, mixing and conditioning methods

Several empiric tests were implemented to optimize the homogeneity and reproducibility ofthe various admixtures considered (with a view on their transposition to an industrial scale).The storage, segregation, dehydration and attrition issues were also studied.Step 7: Study of emplacement methods

Various means of emplacing some powder in a pre-mix of pellets were tested, having again inmind their transposition to an industrial scale.Step 8: Optimizing of mixture components

Two manufacturing processes (tracks) were followed: substitution of an ordinary bentonitepowder by either a powder obtained (after screening) from 7mm OD crushed pellets or from32 mm OD crushed pellets.Step 9: Selection of 3 admixture formulations for the metric scale emplacement tests

At the end of all this empiric testing campaign, 3 admixture formulations were deemed in linewith the requirements and were selected for their eligibility to metric scale emplacement tests(preceding the full scale emplacement test).Step 10: Editing of a final report related to an admixture formulation adapted to FSS

The above testing steps paved the way to a definition of an admixture formulation (to be firsttested and validated at a metric scale) adapted to the construction of the FSS swelling claycore (at scale 1:1).Note:All the iterations related to this methodology are not detailed in the document. The mainoutcomes only are summarized.


4. Preparatory work for the FSS clayish material definition

4.1 Materials used in the lab testsThe following materials were supplied for the purpose of lab testing:

· Bentonite powder WH2, brand name “GELCLAY WH2”,· Pellets 7 mm OD, brand name “EXPANGEL 7”,· Pellets 28 mm OD, brand name “EXPANGEL 28”,· Powder made of crushed pellets 7 mm OD, called hereafter “CP”.Note:

The use of some 28mm OD pellets instead of 32mm OD pellets was justified by the non-availability of the 32mm OD pellets compacting machine (which was still a prototype underdevelopment at time of lab test campaign start-up). The 28mm OD pellets were producedwith the same WH2 powder (dried at 5 % of water content) as that used for the final 32 mmOD pellet production.

All the materials and components were delivered in 25 liter buckets (watertight and stored atambient lab temperature) as detailed in Table 4.1.

Table 4.4.11 : Summary of materials used in lab tests

Designation LECBA Reference LECBA Date of receipt Quantity(kg)

Powder WH2 standard RE12-012 30/08/2012 200

Pellets 7 mm OD RE12-013 30/08/2012 200

Pellets 28 mm OD RE12-015 28/09/2012 400

CP (Crushed pellets powder) RE12-016 11/10/2012 23

4.2 Initial characterization of materials

4.2.1 Water contentThe measured water content of each material is given in Table 2. All measures were made bydry heating the material concerned at 105°C during 24 hours.

Table 4.2 : Water content of materials

Material Water content (%)

Powder WH2 standard 2.3

Pellets 28 mm OD 5.1

Pellets 7 mm OD 4.9

CP (Crushed pellets powder) 4.9


4.2.2 DensityThe “bulk density” was measured (by weighing a predetermined volume of pellets or powderor CP) thanks to the polycarbonate cell (Figure 4.1), while the “(apparent) dry density” wasmeasured thanks to the “wax method”. Table 3 summarizes the results.

Figure 4.1: Measurement of bulk density with 28 mm OD pellets

Table 4.3: Density of materials tested

MaterialBulk

Density(Mg/m3)

Dry BulkDensity(Mg/m3)

AverageMass (g) Density

(Mg/m3)Dry Density

(Mg/m3)

Powder WH2standard 1.084 1.060 - - -

Pellets OD 28mm 1.182 1.124 26.329 2.061 1.961

Pellets OD 7 mm 1.268 1.209 0.461 2.096 1.998

CP (Crushedpellets powder) 1.326 1.264 - - -

Note: For the WH2 powder, the grain density equaled 2.78 Mg/m3.

4.2.3 GranulometryA granulometry spectrum curve was plotted on the WH2 powder (reference LECBA RE12-012). The corresponding results are illustrated in Figure 4.2, confirming pretty well the WH2characteristics furnished by the vendor (LAVIOSA MPC, another member of the GME).


0

10

20

30

40

50

60

70

80

90

100

0.000 0.500 1.000 1.500 2.000 2.500 3.000 3.500

%

maille tamis (mm)

% refus cumulé

% refus cumulé

0

10

20

30

40

50

60

70

80

90

100

0.000 0.500 1.000 1.500 2.000 2.500 3.000 3.500

%

maille tamis (mm)

% passant cumulé

% passant cumulé

Figure 4.2 : Granulometry of bentonite WH2 powder - Reference LECBA RE12-012

4.3 Description of test equipment

4.3.1 Polycarbonate test cellThe polycarbonate test cell (ID 390mm, H 500 mm) was fabricated (Figure 4.3 shows thepolycarbonate cell “as built”) in order to:

· Accurately measure the masses and volumes concerned by the tests,· Reduce as much as possible the « side effects » by providing a significant test volume

(the cell ID is some 12 times the pellet diameter),· Visualize how the materials behave during filling operations and how the “piling” and

“bridging” structures are formed in the mix emplacement,· Take photos,· Benefit from an easy assembly and dismantling device, considering the numerous tests

implemented.

Figure 4.3 : Photo of polycarbonate cell during its commissioning


For the mass measurement, the cell was disposed on a 160 kg capacity scale with an accuracyof some ±50 g (0.03%). For the volume measurement, a 10 mmm thick aluminum lid washorizontally positioned on the top of the materials: the remaining height between the upperpart of the cell and the lower part of the lid was measured with an accuracy of some 1/20 mm.

4.3.2 Mixer and bucketThe mixer used for the tests was a concrete mixer equipped with a fiber glass tank, a speedvariator and a removable lid (used to prevent dust spill).

Following the mixing of the bentonite components, the admixture was poured into thepolycarbonate cell, thanks to a tilting bucket. Figure 4.4 shows the mixer and the bucket.

Figure 4.4: View of mixer, polycarbonate cell and tilting bucket used

4.3.3 Steel ballsIn the waiting of the first experimental bentonite pellets (28 mm OD), an assortment of steelballs, in various diameters was supplied, in order to evaluate the side effects, the pilingstructure and the compacity of the ball arrangement inside the polycarbonate cell. Table 4.4presents the data (an average of 10 measures) related to the steel balls as well as the “inter-ball” void ratio. Figure 4.5 shows the evolution of the “inter-ball” void/space ratio as afunction of the ball diameter.

Table 4.4 : Steel ball characteristics vs inter-ball void/space ratio

Steel ball OD (mm) 25 31.75 34.925 39 44.45

Mass (g) 63.797 130.200 174.324 234.424 358.173

Volume (cm3) 8.181 16.758 22.305 31.059 45.985

Density (g/cm3) 7.798 7.769 7.815 7.837 7.789

Average void ratio (%) 41.86 42.63 43.93 45.37 45.57


40

41

42

43

44

45

46

47

48

49

50

0 5 10 15 20 25 30 35 40 45 50

Vide

inte

rbill

es(%

)

Diamètre des billes (mm)

Billes acier

Figure 4.5: Evolution of the “inter-ball” space ratio as a function of the ball diameter

4.4 Tests with bentonite pellets onlyThe tests were carried out with 28mm OD pellets and 7mm OD pellets (cf. Figure 4.6) todetermine the effective “inter-ball” ratio with the real material. Figure 4.7 shows thecomparison obtained at the end of the tests.

Figure 4.6: Emplacement tests with 28mm (left) and 7 mm (right) OD pellets


pellets 28 mm

pellets 7 mm

30

32

34

36

38

40

42

44

46

48

50

0 5 10 15 20 25 30 35 40 45 50

Vide

inte

rbill

es(%

)

Diamètre des billes (mm)

Figure 4.7: Comparison of the inter-pellets void as a function of the pellet diameter

4.5 Tests with binary mixturesFive binary mixtures were tested, with the following (28mm OD) pellet / (standard WH2)powder ratios: 60/40, 65/35, 70/30, 75/25, 80/20 (the ratios are calculated on the basis of drymass of each component), as shown in Figure 4.8. The tests were then repeated with abentonite powder made of crushed pellets (CP).

Figure 4.8: Emplacement tests with binary mixtures

The comparison of the results obtained showed (cf. Table 4.5) that the most relevant mixturewas that formulated with a ratio 70/30 and a powder made from crushed pellets (CP 0-5 mm)inserted in 10 layers in the inter-pellets voids.

It was inferred that with an effective 32 mm OD pellet, the emplaced density could be easilyachieved and that the relevant mixture was the one identified during these definition tests,provided the pellet diameter change (from 28mm to 32mm) was made effective.


Table 4.5: Results obtained on binary mixture with 28mm OD pellets and crushed powder

TestMass ofpellets

(g)

Mass ofpowder

(g)

%pellets

%powder

Totalmass(g)

Height(mm)

Volume(cm3)

ρemplaced

(g/cm3)

ρ dryemplaced

(g/cm3)

Volumeof void(cm3)

%void

Totalporosity

(%)

Pellets 28mm OD +crushed

powder in10 layers

38960 17380 68,59 31,41 56340 293,64 34796 1,619 1,555 2790 8,02 44,08

4.6 Tests with the real componentsOnce the production of 32mm OD pellets started, it was possible to repeat the series of testswith the real components and to make the necessary fine tuning formulation adaptation inorder to reach the optimum method.The following materials (cf. Figure 4.9) were delivered to LECBA for the purpose:

· Some 120 kg of pellets 32 mm Expangel SP32, referred at LECBA as n° RE13-005,· Some 100 kg of crushed powder 0-2 mm, referred at LECBA as n° RE13-006,· Some 100 kg of crushed powder 0-5 mm, referred at LECBA as n° RE13-007.The granulometry curve of crushed powder RE13-007 is provided in Figure 4.10.

Figure 4.9 : Pellets RE13-005 – Crushed powder RE13-006 and RE13-007

0

10

20

30

40

50

60

70

80

90

100

0.000 1.000 2.000 3.000 4.000 5.000 6.000

%

maille tamis (mm)

% refus cumulé

0

10

20

30

40

50

60

70

80

90

100

0.010 0.100 1.000 10.000

%

maille tamis (mm)

% refus cumulé

Figure 4.10 : Granumometry of crushed powder RE13-007Figure 4.11 shows an example of one of a series of tests where the pellets are first introduced,and later the crushed powder is inserted. The Table 4.6 summarizes the best results obtainedwith 3 formulation combinations.


Figure 4.11: Backfilling test with 32mm OD pellets and crushed powder RE13-007

Table 4.6: Results obtained with the best binary mixtures (32mm OD pellets & CP)

Test %pellets

%powder

ρemplaced(g/cm3)

ρ dryemplaced(g/cm3)

Voidrate(%)

Total

Porosity (%)

Pellets 32 - 4 x 3layers + crushed

powder 0-268.2 31.8 1.663 1.599 5.04 42.50

Pellets 32 - 10layers + crushed

powder 0-270.3 29.7 1.631 1.569 7.85 43.57

Pellets 32 4 x 3layers + crushed

powder 0-568. 9 31.1 1.644 1.581 4.77 43.14

4.7 Selection of the bentonite formulationOther iterations of tests were carried out with the above formulations, taking into account,other criteria such as industrial feasibility (homogeneity), logistics (packing, transport, andconservation), use (integrity, production of dust).

The water content of the raw material and the compaction pressure of the pelletmanufacturing machine were also adjusted to find the best possible compromise.

At the end, it was decided that bentonitic admixtures with 70% of 32 mm pellets mixed onelayer at a time with some 30% of crushed powder CP 0-5mm would be the optimum solution,but that the emplacement method had to be first reproduced/adapted at a metric scale beforeimplementing the full scale (1:1) backfilling test in the FSS drift model.


5. Mock-up testing and emplacement machine design

5.1 IntroductionIn parallel with the laboratory studies, a mock-up metric-scale set-up was developed fortesting the selected bentonite admixtures and the backfilling device concept. It was deemed“necessary and careful” to test first at an intermediate scale (metric), the behaviour of thebentonitic mixes selected, before later going at full scale (decametric), since the manualemplacement methods carried out at a decimetric scale in lab could not be easily transposed(“from scratch”) to an industrial application at a decametric scale.

The test mock-up set-up is illustrated in Figure 5.1. The emplacement device is based on theuse of two hoppers (one for pellets, one for CP) and two conveying augers for the respective

conveying of pellets and powder (

Figure 5.21).


Figure 5.1: Metric emplacement test set-up


Figure 5.2: Mock-up tests of the bentonite pellet-powder mix using conveying augers

5.2 OutcomesGlobally, the use of the mechanical emplacement device selected turned out to be relevant.The conveying augers managed to efficiently backfill the mock-up concrete pipe (cf. Figures5.2 & 5.3) with the bentonitic mix and could also provide some “backfill pressure” highenough to fill the drift model summital voids (as simulated in Figure 5.4).

However, the obtained bentonite density was less than the specified value of 1620 kg/m3

which was manually obtained in laboratory. The best values of the obtained density in thismock-up test were 1510 kg/m3 with the powder auger above the pellets auger and 1470 kg/m3

with the two augers side by side. Besides, the second position (side by side) turned out to methe most adapted one to exert an efficient backfilling pressure. This second position was alsothe one creating the most segregation between the emplaced components.

The main reason for these lower density values was deemed to be the breakage of somepellets during the handling process resulting in closure of inter-pellet spaces and preventingthe powder from accessing some voids. As a result, mechanical resistance tests wereintroduced to measure the “hardness” of the pellets (resistance to erosion or breakage due to acompacting effort as envisaged inside the screw conveyor pipe) at the pellet productionworkshop and (as a measure of quality control) at the FSS site before emplacement.

A conveyor belt was also added to minimize the length of the passage of pellets inside theconveying augers, thus minimizing pellet erosion/attrition and breakage (cf. Figure 5.5). Thissolution was effective and incorporated in the design of the full scale emplacement machinefor the backfilling of the swelling clay core in the drift model.


Figure 5.3 : Progressive backfilling of the mock-up pipe

Figure 5.4 : Backfilling pressure enabled to fill the summital voids


Figure 5.5 : Incorporation into the emplacement test set-up of a pellet conveying belt

Besides, an evaluation of the relationship between dry density and swelling pressure for WH2undertaken in parallel with the metric emplacement tests described above showed that with adry density of 1500 kg/m3 only the swelling pressure obtained at saturation of material wouldbe around 4 MPa.

This pressure was finally considered by Andra to be sufficient (as far as the permeabilitytarget was still preserved and the (host formation/argillite) EDZ self-sealing was stilleffective), and the FSS swelling clay core construction specifications were accordinglymodified so that the required average dry density in the clay core would be 1500 kg/m3

instead of 1620 kg/m3 (as originally specified).


6. Characterization of the bentonite mix

6.1 IntroductionEven if the relation between the emplaced density value of the selected bentonitic mix andthe swelling pressure was first established by experience (i.e. by extrapolating data fromscientific literature on bentonite MX80), it was necessary to confirm that the selectedbentonitic material formulation effectively reached the specified swelling pressure, i.e. some4 MPa as defined at the end of the metric emplacement tests.

This aspect was explored in 2 phases and at 2 different scales:· First, an oedometric test was carried out in lab on the FSS admixture in a test cell (at a

decimetric scale), over a few months; this part is reported below.· Then the REM test, carried out at the Bure/Saudron Technical Center (at a metric test),

with the same FSS admixture, over a few decades (20-30 years). This experiment isongoing and described in Conil et al. (2015) - DOPAS Work Package 4, Deliverable D4.2(“Report on Bentonite Saturation Test (REM)”).

6.2 Experimental protocol and resultsThe FSS mix sample (composition is shown in Table 6.1) was placed inside a speciallymanufactured oedometric cell (cf. Figure 6.1) and progressively saturated. Saturation tooksome 181 days and maximum swelling pressure was reached after that lapse of time.

The experiment was then dismantled. The saturated bentonitic material appeared asthoroughly homogenized, without possible visual distinction between the pellets and thecrushed powder (cf. Figure 6.2).

Table 6.1: Characteristics of FSS mix sample submitted to oedometric test.

Mass(g)

DryMass(g)

Proportion(%) Height(mm) Volume

(cm3)

ρemplaced(g/cm3)

ρ dryemplaced(g/cm3)

PoreVolume

(cm3)

TotalPorosity

(%)Sr (%)

PelletsRE14-002 3258.6 3121.0 69.,6

65.65 2970 1.577 1.510 1356.3 45.67 14.63

CP

RE14-0031425.6 1364.9 30.,4

Total 4684.2 4485.8 100

The measured pressure obtained after complete saturation was 3.88 MPa, i.e. slightly belowthe 4 MPa target. This value was however deemed acceptable to proceed with:

· The backfilling of the FSS swelling clay core inside the drift model,

· The launching of the REM experiment.


Figure 6.2.1: Illustration of oedometric test cell backfilling with the FSS mix


Figure 6.2.2 : Dismantling of test cell and view of saturated/homogenized FSS mix


7. Conclusion

At the end of the swelling clay core definition, metric emplacement tests andcharacterization, it was concluded that:

· The mix recipe was robust, and that the fabrication of the material manufacturing forFSS could be launched;

· The swelling pressure measured in lab was close enough to the 4 MPa target toinitialize the REM experiment;

· The design of the FSS backfilling machine could be frozen and its fabrication ordered(incorporating the pellet conveying belt).

The subsequent swelling clay core construction is reported in Bosgiraud et al. (2016) -DOPAS Work Package 3, Deliverable D3.12 (“Report on construction of FSS swelling claycore”).


8.References

· Andra (2013) - The Cigéo Project – “Meuse/Haute-Marne Reversible GeologicalDisposal Facility for Radioactive Waste” - Project Owner File Public Debate of 15May to 15 October 2013.

· White et al. (2014) - DOPAS Work Package 2 - Deliverable D2.1 (“Design Bases andCriteria”).

· DOPAS (2016a) - WP2 Final Report: Design Basis for DOPAS Plugs and Seals.DOPAS Work Package 2, Deliverable D2.4, Version 1.

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.1 (“FSSconstruction summary report”).

· Bosgiraud et al. (2013) - DOPAS Work Package 3, Deliverable D3.2 (“FSS tunnelmodel design report”).

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.3 (“Report onclayish material definition for FSS”) – combined with D3.7.

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.4 (“Report on lowpH concrete formulas for FSS”) – combined with D3.8.

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.5 (“Lab testReport on the performance of the clayish material for FSS”) – combined with D3.7.

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.6 (“Lab testReport on the performance of low pH concrete for FSS”) – combined with D3.8.

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.7 (“Test report onFSS metric core emplacement, with clayish materials definition for FSS andLaboratory work on the performance of clayish materials for FSS”).

· Bosgiraud et al. (2013) - DOPAS Work Package 3, Deliverable D3.8 (“Test report onFSS cast in box concrete with low pH concrete formulas for FSS and Laboratorywork on the performance of low pH concrete for FSS”).

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.9 (“Test report onFSS test panel for shotcrete”).

· Bosgiraud et al. (2013) - DOPAS Work Package 3, Deliverable D3.10 (“FSS Driftmodel construction report”).

· Bosgiraud et al. (2014) - DOPAS Work Package 3, Deliverable D3.11 (“FSS castconcrete plug construction report”).

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.12 (“Report onconstruction of FSS swelling clay core”).

· Bosgiraud et al. (2016) - DOPAS Work Package 3, Deliverable D3.13 (“Report onshotcrete plug construction”).

· DOPAS (2016b) - DOPAS Work Package 3 - Deliverable D3.30 (“WP3 FinalSummary Report: Summary of, and Lessons Learned from, Design and Constructionof the DOPAS Experiments”).


· Noiret et al. (2016) - DOPAS Work Package 4, Deliverable D4.1 (“Report onQualification of Commissioning Means”).

· Conil et al. (2015) - DOPAS Work Package 4, Deliverable D4.2 (“Report onBentonite Saturation Test (REM)”).

· DOPAS (2016c) - DOPAS Work Package 4 Deliverable D4.4. WP4 IntegratedReport. Summary of Progress on Design, Construction and Monitoring of Plugs andSeals.

· Bosgiraud et al. (2016) - DOPAS Work Package 4, Deliverable D4.8 (“FSSExperiment Summary Report”).

· DOPAS (2016d). DOPAS Work Package 5, Deliverable D5.10. WP5 FinalIntegrated Report. DOPAS Project Deliverable D5.10.

· DOPAS (2016e) - DOPAS Final Project Summary Report. DOPAS Work Package 6,Deliverable D6.4. DOPAS Project Deliverable D6.4.